the effects of school resource officers on feelings of
TRANSCRIPT
The Effects of School
Resource Officers on
Feelings of School
Connectedness and Safety Word Count: 5443
I. Introduction
The implementation of School Resource Officers (SROs) in public schools, as well as
their effects on students, are becoming increasingly important. Although School Resource
Officers (SROs) have been implemented in U.S. high schools since the mid-1900s, school
crime was not yet a national concern at the time of the program’s conception. According to
sociologist Dr. Ben T. Brown, school crime was once thought to be a problem of exclusively
impoverished schools (Brown 2006). The shooting at Columbine High School challenged
this belief, as it was considered the deadliest school massacre in the nation’s history (Brown
2006). Events of school violence both in the past and present have heightened public concern
of school violence, reflected in the increased implementation of SROs as a prevention
measure.
Despite their prevalence, the effectiveness of SROs on improving school safety and
perceptions of safety have been widely debated. After the school shooting at Marjory
Stoneman Douglas High School in Parkland, Florida, many people, including President
Trump, have pushed for SROs in every school as violence prevention measures. However,
“The school in Parkland, Fla. had a school resource officer on duty during the shooting. The
Broward Sheriff said the officer remained outside in a defensive position at the time of the
shooting” (Corley 2018). The continuous debate surrounding school safety has evidently
raised questions as to the effectiveness of SROs after their failures in preventing attacks.
Furthermore, the use of SROs in public questions has called into question their further
effects on students. Research has focused on SROs effects on perceptions of safety and
feelings of school connectedness. A study conducted by Matthew T. Theriot and John G.
Orme, administered a school safety survey to a Southeastern United States school district
(Theriot, Orme 2016). It resulted that students who had more positive attitudes toward SROs
on campus felt safer (Theriot, Orme 2016). Although positive attitudes toward SROs have
been related to boosted perceptions of safety, they've also been found to decrease school
connectedness, or students' feelings of connection and belonging to a school. SROs have
been extensively researched to reveal their effects on students, shifting the focus outside of
their main function as security measures.
In order to contribute new insight to the current body of research, the present study will
use previously established methods in a student body that differs ethnically and
socioeconomically. By administering a self-report survey developed by credible researchers
and screening potentially exaggerated scores, the present study will serve to address the
question: How do perceptions of school resource officers on high school campuses affect
students’ perceptions of safety and school connectedness?
II. Literature Review
School Resource Officer Programs
Since the mid-1900s, school resource officer programs have been implemented in schools
across the U.S., employing officers (SROs) to patrol and investigate crimes on school properties.
However, the number of SROs increased dramatically during the 1990s, and by the year 2000,
nearly half of the U.S.’s police departments had assigned full time officers to serve as SROs at
schools (Brown 2006). According to a National Assessment of School Resource Officers by Jack
McDevitt and Jenn Panniello, the goal of the SRO program is to ensure safety (McDevitt,
Panniello 2005). SROs act as counselors and teach students within their school community in
such a way that helps prevent crime (McDevitt, Panniello 2005). While these SROs aim to deter
school crime, many studies have undermined their effectiveness. In a study by Arrick Jackson in
2002, a survey was administered to high school juniors and seniors, attending either an SRO or a
non-SRO school (Jackson 2002). This survey asked their opinions of the seriousness of
delinquency, perceptions of the SRO, their perceptions of being identified if committing
delinquent acts, interactions with SROs and their liking of the SRO (Jackson 2002). Although
they found that SROs had nearly no effect on any of the scales, students attending a high school
with an SRO believed they were less likely to be detected when involved in delinquent activity
(Jackson 2002). Jackson’s study found that SROs were mostly effective in deterring assault on
campus, but did not show any further effects on other crimes or the student perceptions of crime
(Jackson 2002). Studies like Jackson’s highlight instances in which SROs have not been
effective in deterring crime and ensuring safety in a school community.
SROs and Perceptions of Safety
Due to the increased implementation of SROs, studies have been conducted to explore
their effects on students and faculty, such as their perceptions of safety. In a study by Theriot and
Orme, a survey administered to a Southeastern United States school district showed that students
who had more positive attitudes toward SROs on campus felt safer (Theriot, Orme 2016).
Additionally, they concluded that student experiences with school violence, rather than
interactions with SROs, had a more significant impact on their feelings of safety (Theriot, Orme
2016). Another study by McDevitt and Panniello focused on the factors that affect both students’
comfort reporting crimes to SROs and their perceptions of safety in their schools (McDevitt,
Panniello 2005). Through a survey administered to middle and high school students, they found
that knowing the SROs’ name as well as having conversations with them could positively affect
whether the student felt comfortable with reporting crimes to their SRO (McDevitt, Panniello
2005). In this study, perceptions of safety were most impacted by the level of neighborhood
crime, past victimization at school, their comfort reporting crimes, having a positive opinion of
SRO, and gender (McDevitt, Paniello 2005). Since there have been various interpretations of the
effects SROs have on perceptions of safety, it is imperative that further research is conducted to
make these effects clearer.
Student Perceptions of School Security Measures and Safety
Understanding the factors that construct one’s perceptions of safety provide insight into
how SROs might affect a students’ perceptions of safety. One of these factors is the presence of
security measures. In a study by Bosworth and colleagues in 2011, when randomly selected
students and teachers were asked “What makes a school safe”, they replied within 3 categories:
physical characteristics and safety features, organization and school discipline, and school
staffing and relationships that enhance feelings of safety with a sense of caring and community
(Bosworth et al. 2011). From these three, students ranked tangible security equipment higher
than school staffing as a contributor to feelings of safety (Bosworth et al. 2011). Both students
and faculty found that school organization and discipline, as well as school staffing and
relationships, made them feel safer at school (Bosworth et al. 2011). This study showed that both
students and staff felt safest in schools where they perceived the adults to be caring and helpful,
and that rules were clear, consistent, well communicated, and consistently applied (Bosworth et
al. 2011). These findings suggest that adults and students share factors that construct their
perceptions of safety. Another study by Ben Brown and W.M. Reed Benedict in 2005 focused on
student perceptions of school security measures and the effect on their feelings of safety (Brown,
Benedict 2005). A survey administered to students of the Brownsville Independent School
District showed that while students are supportive of many of the safety measures, those
employing “hard control” have a limited impact on school crime overall (Brown,Benedict 2005).
These hard control methods, such as video surveillance, are similar to the physical security
measures studied by Bosworth and colleagues to which they referred to as physical
characteristics and safety features. However, there are contradictory results between the two
studies since the students surveyed by Bosworth considered these measures to be more important
to their feelings of safety, but the students surveyed by Brown and Benedict found the measures
to be ineffectual in deterring crime. While security measures are an essential part of student
perceptions of safety overall, the conflicting opinions lead researchers to believe there are other
relevant factors.
The study “Beyond Guns, Drugs and Gangs” by Skiba and colleagues in 2004 addresses
these relevant factors. The researchers surveyed students, staff, and parents about school safety
(Skiba et al. 2004). The most important factors of their feelings of safety for all three groups
were the connection or climate of the school, incivility and disruption, personal safety, and
delinquency or major safety (Skiba et al. 2004). This study showed that school connection and
climate are more important than factors of delinquency or physical safety measures in predicting
students’ overall feelings about school safety in the locations studied by the researchers (Skiba et
al. 2004). Additional studies have also indicated that experiences with different forms of
aggression contribute to students’ perceptions of their school climate as well as their feelings of
safety at school. A study conducted by Sara Goldstein and colleagues called “Relational
Aggression at School: Associations with School Safety and Social Climate” surveyed middle
school and high school students in metropolitan Detroit provided insight into how students’
experiences at school affect their perceptions of safety (Goldstein et al. 2007) . Their survey
results indicated that high exposure to relational (verbal) aggression made students feel less safe
at school and have negative overall experiences at school, impacting their perceptions of their
social climate negatively (Goldstein et al. 2007). Since school climate and connection have been
found to have an impact on student perceptions of safety, the impact SROS have on those two
factors must also be addressed.
SROs and Feelings of School Connectedness
Another focus taken on the effects of SROs are their influence on the school
connectedness perceived by students. In a study by Matthew T. Theriot in 2016, the researcher
administered a survey to a school district in Southeastern United States about school safety
(Theriot 2016). The survey collected the amount of times students reported interacting with
SROs, and the students’ responses to questions about school connectedness and climate. These
questions were compiled into distinct scales to construct scale scores (Theriot 2016). The higher
the participants scored on either the school connectedness or SRO perceptions scales, the more
negative feelings of school connectedness or more positive opinions of the SRO were observed
(Theriot 2016). The survey showed that increased student interaction with SROs caused more
positive attitudes toward the SROs, but lower school connectedness. This was explained by
Theriot to possibly be a result of the increased interactions with SROs drawing attention to
school crime, which can generate worry or violence among students, and contribute to the
students feeling less connected and comfortable at school (Theriot 2016). By considering how
SROs affect school connectedness, there is a better understanding of how they affect students’
perceptions of safety overall, considering the conclusion the study by Skiba and colleagues that
identified it as a relevant factor to how safe students feel at school.
Validity of Self Report Survey Method
The majority of studies on student perceptions of safety have used self-report surveys to
identify the effects of SROs on students and faculty. Due to the reliance on this method, many
have assessed the validity of self-report surveys to gauge perceptions of safety in response to
SROs. In a study by Cornell and Loper, it was demonstrated that by screening surveys for
incomplete or careless responses, surveys can be essential in gaining insight into the nature and
extent of high-risk behavior in schools, and effectively help school psychologists and other
educators in developing violence intervention and prevention programs (Cornell, Loper 1998).
These methods established by Cornell and Loper were used in Theriot’s various studies on
school resource officers, including in his work in conjunction with John G. Orme. This was done
to exclude careless reporting or surveys that showed high levels of victimization. While Theriot
and Orme explicitly used these screening methods to ensure the validity of their survey
responses, much of the research about student perceptions of safety and SROs does not. For
example, the study “Police-School Officers and students’ perceptions of police and offending”
by Arrick Jackson implements the self-report survey method to measure student perception of
delinquency and school-resource officers or police, depending on whether the school they
focused on used an SRO. The survey data indicated that increased interaction with SROs resulted
in more positive opinions of the students’ SRO, but it did not change their opinions of police
outside of their school (Jackson 2002). While this survey was consistent with the results of the
study done by Theriot in 2016, there was no screening process implemented during data analysis,
which is essential in excluding careless responses and improving the validity of the self-report
survey method. This lack of screening proves to be missing in much of the research in this area.
Hypothesis
The mission of this research is to explore the relationships between school resource
officers and high school students’ perceptions of safety and school connectedness. The
researcher of this study hypothesizes that in an environment that differs ethnically and
socioeconomically than previous studies, school safety survey results will support previous
studies that found SROs to affect perceptions of safety positively and student’s feelings of school
connectedness negatively. This would be conveyed through increased feelings of safety but
decreased levels of school connectedness in response to more interactions with or more positive
opinions of the SRO.
III. Methods
A. Population: The school included the present study, GHC, is a large
sized co-ed public high school in a suburban setting. There are 4,480
attending students, with an ethnically diverse population, outlined in
Table 1. GHC’s population contrasts with this study’s foundational
sources due to differences in regional settings and ethnic demographics.
These differences may prove to create a difference between this the
results of this study’s population and of its foundational sources’
population.
B. Survey Alignment: In order to align with previous methods implemented in studying
student perceptions of SROs and safety, a researcher expert was contacted. Matthew T.
Theriot and John G. Orme, professors of the College of Social Work at the University of
Tennessee, have previously researched student perceptions of SROs through survey
distribution. In order to closely align the present research to previously established
Race/
Ethnicity
Percentage
(%)
Hispanic 40
White 25
Asian 18
Filipino 9
African
American 4
American
Indian/Alaska
Native <1Pacific
Islander <1
Declined To
State <2
Source: GHC 2017-18 School Profile
Table 1: GHC Ethnic
Demographics
methods, Theriot was contacted through email and he provided the survey used in his and
Orme’s study. This survey was then included in the present study to be distributed after a
stratified random sample of GHC student was selected. From this survey instrument, the
questions making up each scale regarding school connectedness, school safety
perceptions, violence experienced at school are described in Table 2.
TABLE 2: Instrument Table
Questions Measurement Scale Source
Demographic Questions Assorted
What gate did you enter the school from the morning you were contacted?
Flagpole/Zelzah, Zelzah Teacher Parking Lot, Service Road
(Boy's PE side), Hiwatha Parking Lot, Girls PE Side Service
Road, J Gate, Kingsbury (main entrance) self-defined
What is your gender? Male, Female self-defined
What grade are you in at schoool? Freshman, Sophmore, Junior, Senior self-defined
How many years have you been a student at this school? 1 (this is my first year), 2, 3, 4 ,5 or more Theriot, Orme 2016
How many good friends do you have at school?
None, I have one good friend at school, I have 2-3 good friends
at school, I have 4-5 good friends at school, I have 6 or more
good friends at school
School Connectedness Scale Statements Likert Scale
I am learning a lot at this school.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree Theriot, Orme 2016
School rules seem fair to me.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
Teachers work hard to make every student successful.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
I feel that I belong at this school.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
I do not try hard in school.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
In general, I like school.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
I don’t care what teachers think of me.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
I am proud of this school.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
I do not feel that I can tell a teacher, principal, or other adult at school if I have a problem.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
Students enjoy learning here.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
Student School Safety Perceptions Statements Scale Likert Scale
Overall, I feel safe at school. 1=strongly disagree; 3=neutral; and 5=strongly agree Theriot, Orme 2016
I feel safe in the school hallways. 1=strongly disagree; 3=neutral; and 5=strongly agree
I feel safe in the cafeteria. 1=strongly disagree; 3=neutral; and 5=strongly agreeI feel safe going to school in the morning and coming home from school in the
afternoon. 1=strongly disagree; 3=neutral; and 5=strongly agree
I feel safe in the bathrooms at school. 1=strongly disagree; 3=neutral; and 5=strongly agree
I feel safe in my classrooms. 1=strongly disagree; 3=neutral; and 5=strongly agree
In general, I do not trust the police. 1=strongly disagree; 3=neutral; and 5=strongly agree
Having a police officer at my school everyday would make me feel safer. 1=strongly disagree; 3=neutral; and 5=strongly agree
Having metal detectors at my school would make me feel safer. 1=strongly disagree; 3=neutral; and 5=strongly agree
Having surprise locker checks at my school would help me feel safer. 1=strongly disagree; 3=neutral; and 5=strongly agree
In general, I like the police. 1=strongly disagree; 3=neutral; and 5=strongly agree
Gangs are a big problem at my school. 1=strongly disagree; 3=neutral; and 5=strongly agree
Student Experience with Violence at School Likert Scale
During this school year, how often have you been teased, called names, or insulted
by another student while at school?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times Theriot, Orme 2016
During this school year, how often have you had a physical fight with another
student while at school?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how often have you had an argument with another student
while at school?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how often have you been bullied by another student while
at school?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how often have you been hit, kicked, grabbed, or otherwise
physically grabbed, or otherwise physically harmed by a boyfriend, girlfriend, or
anyone that you were dating?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how often have you been teased, called names, insulted, or
attacked at school because of your skin color, race, religion, or where your family
came from?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how often has another student threatened to physically
harm you while at school?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
School Violence Students Have Seen Likert Scale
During this school year, how many times have you seen students smoke cigarettes or use
tobacco at school?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times Theriot, Orme 2016
During this school year, how many times have you seen a knife at school (not including a
cafeteria knife)?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you seen a gun at school (not including police
officers)?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you seen students with drugs or alcohol at
school?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you seen a student tease
or insult another student (including name-calling)?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you seen a student bully another student?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you heard a student threaten to physically harm
another student?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you seen violence between two students who are
dating each other?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you seen two or more students have a loud
argument?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you seen two or more students get in a physical
fight?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
During this school year, how many times have you seen a student tease, insult, or attack
another student because of skin color, religion, or where they came from?
1=never, 2=once or twice, 3=about once a month, 4=2–3 times a
month, 5= about once a week, and 6= several times
Attitudes About SROs Scale Statements Likert Scale
I like having a police officer at my school everyday.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree Theriot, Orme 2016
I feel safer with the police officer at my school.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
The police officer at my school does not treat all students fairly.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
In general, the police officer at my school does a good job.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
The police officer at my school has a good relationship with the students.
= strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
The police officer l does a good job of stopping violence at my school.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
The police officer does a good job of stopping students from selling or using drugs or alcohol at
school.
5= strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
The police officer at my school does not make me feel better about the police in general.
5= strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
The police officer at school will help students who need help
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
Most students at my school don't like the police officer.
5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly
disagree
Number of Interactions with SRO Assorted
During this school year, how many times have you interacted with the police officer at your
school (including things like talking to him or her, asking for or getting help from the police
officer, and getting in trouble)? None, 1-2 times, 3-4 times, 5-6 times, 7 or more times Theriot, Orme 2016
During this school year, has the police officer at your school led a class or given a presentation
that you attended? Yes , No
Source: My December 2018 GHC Study
C. Sample Selection: In order to make each student equally likely to be chosen to
participate in the study, a stratified random sample was done every day before school for
3 weeks. Each of the seven gates on campus has a certain percentage of students who
come through it. In order to perform a stratified random sample, students were randomly
chosen and asked to participate in the study. The gate strata, and the percentage of
students included are outlined in Table 3. With the help of two business statistics
students, students coming into school through each gate were asked if they wanted to
participate in the study. All GHC students have access to a Chromebook and Gmail
account, providing an easily accessible method of distributing the survey. Through this
process, 125 students were able to participate in the present study. While there was a
sample of 125 students from all of the gates, a further random sample of 100 participants
out of the 125 was chosen using Excel technology in order to facilitate the data
processing and assure the gate proportions were met for each stratum.
School EntrancePercentage of Students
(%)
Flagpole (Zelzah) 28
Kingsbury 24
Zelzah Teacher
Parking Lot22
Service Road
West11
Hiawatha Lot 7
Service Road East 5
J-Gate 3
Source: My December 2018 GHC Study
Table 3: Stratified Random Sample
D. Implementation: The survey implemented in this study was mechanically distributed
through school-issued Chromebooks and Google Forms. Each time a participant was
identified at the gate, the survey was sent to them digitally through an email with a
Google Forms link. Since each student has access to the technology implemented, every
student in the population was equally likely to participate. The data from each survey was
processed through Excel Toolpak functions such as histogram constructions or univariate
data calculations.
E. Survey Screening Alignment: A survey screening process was implemented to validate
the results, removing any incomplete or exaggerated surveys. This was done in alignment
with the study by Cornell and Loper demonstrated that by screening surveys for
incomplete or careless responses, surveys can be essential in gaining insight into the
nature and extent of high-risk behavior in schools (Cornell and Loper 1998). Similarly,
the study “Assessing the Reliability and Validity of Student Self-Reports of Campus
Violence” by Jennifer Rosenblatt and Michael Furlong indicated that screening for
incomplete responses and designing procedures that exclude illegitimate answers allow a
more accurate study of school safety and campus violence (Rosenblatt and Furlong
1997). This screening process, implemented through the Google Surveys technology, is
integral to the validity of the self-report survey used in this study, due to the many of the
questions’ subject matter of school violence. It is possible that some respondents may
have carelessly filled out the survey and inaccurately portrayed the sample population, so
by screening the results, the present study has aligned itself to previous methods of
survey distribution.
IV. Findings
A. Histograms
62%
38%
Female Male
Source: My December 2018 GHC Study
What is your gender?
26%
21%
26% 27%
Freshman Sophomore Junior Senior
Source: My December 2018 GHC Study
What grade are you in at school?
26%24% 24%
26%
1 (this is my firstyear at this school)
2 years 3 years 4 years
Source: My December 2018 GHC Study
How many years have you been a student at this school?
0 1
17
3943
None I have 1 goodfriend atschool
I have 2-3good friends at
school
I have 4-5good friends at
school
I have 6 ormore goodfriends at
schoolSource: My December 2018 GHC Study
How many good friends do you have at school?
23
5
1
2019
1716
12
2 20 0 0 0
1
16 18 20 21 22 24 26 28 30 32 34 36 38 40 42 44Source: My December 2018 GHC Study
School Connectedness Scores
01
0 0
21
0
4
21
6
9
3
89
14
17
32
6
23
7
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30Source: My December 2018 GHC Study
Safety Perceptions Scale Scores
B. Descriptive Statistics
27
19 19
10
6
3 42 1 1 1 1 0
20 1
3
0 0 0 0 0 0 0
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Source: My December 2018 GHC Study
Experienced Violence Scale Scores
5
4
5
7
9 9
7
0
8
4 4
2
3
7 7
1 1
3
0
2
0
2
1
2
0
1 1 1
0
1
0 0 0 0 0 0
1
0
1 1
11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49
Source: My December 2018 GHC Study
Witnessed Violence Scale Scores
2
6
15
32
1922
2 2
20 22 24 26 28 30 32 34
Source: My December 2018 GHC Study
SRO Perceptions Scale Score
94
6 0 0 0
None 1-2 times 3-4 times 5-6 times 7 or moretimes
Source: My December 2018 GHC Study
During this school year, how many times have you interacted with the police officer at your
school (including things like talking to him or her, asking for or getting help from the police officer,
and getting in trouble)?
5
95
Yes NoSource: My December 2018 GHC Study
During this school year, has the police officer at your school led a class or given a presentation
that you attended?
C. Correlations
D. Regressions
Survey Scale Possible Range Mean score Standard Deviation
School Connectedness 10 to 50 25.29 0.417
Safety Perceptions 6 to 30 22.32 4.37
Experienced Violence 7 to 42 10.11 4.32
Witnessed Violence 7 to 42 20.95 8.35
SRO Perceptions 10 to 50 26.88 2.8
SC Score SP Score EV Score WV Score SROP Score
SC Score 1.00
SP Score 0.43 1.00
EV Score -0.29 -0.28 1.00
WV Score -0.18 -0.23 0.65 1.00
SROP Score 0.19 0.20 -0.09 -0.09 1.00
slight positive slight negative
strong positive
Interactions Class/Presentation SC Score SP Score SROP Score
Interactions 1.00
Class/Presentation -0.06 1.00
SC Score -0.31 0.02 1.00
SP Score -0.01 -0.05 0.43 1.00
SROP Score -0.02 0.12 0.19 0.20 1.00
slight negative slight positive
V. Analysis of Findings
The foundational study by Theriot and Orme utilized Latent Class Analysis (LCA) and
logistic modeling regression. Regression to analyze the survey results. The LCA serve to identify
discrete subpopulations, or latent classes, of similar cases. The logistic modeling regression
implemented the latent classes identified in the LCA in bivariate and multivariate regression. In
order to best align the present study’s analysis method to those of their foundational studies, the
researcher of the present study utilized the descriptive statistics, correlation, and multivariate
logistic regression functions available on Excel technology.
Comparing Descriptive Statistics and Overall Trends
Due to the manner in which this study has aligned itself with the study done by Theriot
and Orme, the survey data descriptive statistics and overall trends were important to highlight
any differences between the two populations. The survey was broken into demographics, then
into multiple scales addressing school connectedness, safety perceptions, violence experienced,
violence witnessed, and SRO perceptions. The survey was concluded with two questions
Safety Perceptions Score Regression
CoefficientsStandard Error t Stat P-value
Intercept 10.57 4.51 2.34 0.02
SC Score 0.37 0.10 3.78 0.00
EV Score -0.21 0.10 -2.13 0.04
SROP Score 0.16 0.14 1.16 0.25
School Connectedness Score Regression
Coefficients Standard Error t Stat P-value
Intercept 15.08 4.22 3.58 0.00
SP Score 0.35 0.09 3.78 0.00
EV Score -0.16 0.09 -1.65 0.10
SROP Score 0.15 0.14 1.09 0.28
indicating levels of interaction with the SRO. Each demographic question and stand-alone scale
was run through descriptive statistics with Excel technology and examined for overall trends.
Demographics
The first questions addressed the participants’ demographics, such as gender, grade, years
spent at this particular school, and number of good friends at school. The gender demographics
were 62 percent female and 38 percent male. For grade level, the proportions of each group were
close to equal, where there were 26% freshmen, 21% sophomores, 26% juniors, and 27%
seniors. This remained true for the spread for years spent at school since it was nearly uniform,
meaning that each group was nearly equal. For the number of good friends at school, more than
half of the respondents (72%) have at least 4 good friends at school. Only small percentage of
students (18%) have less than 4 good friends at school. Compared to the Theriot and Orme study
in which 1,744 students had 2 or more good friends at school (89%), the present study had 99
students with 2 or more good friends at school (99%) (Theriot, Orme 2016).
School Connectedness
The stand-alone scales were also run
through descriptive statistics with Excel
technology to identify any trends. The first survey
scale was after the demographic questions,
addressing the participants’ feelings of school
connectedness with 10 questions. It asked
questions about how fair they believed their school rules to be, work ethic, school pride, sense of
belonging, liking of school, whether the students are learning a lot and their willingness to report
23
5
1
2019
1716
12
2 20 0 0 0
1
16 18 20 21 22 24 26 28 30 32 34 36 38 40 42 44Source: My December 2018 GHC Study
School Connectedness Scores
problems to school staff. Each question within this scale had Likert scale responses where 5 =
strongly agree; 3 =agree, neutral; and 1=disagree, strongly disagree. To create the overall score,
each response was simply summed to one score. The school connectedness scale had a possible
range of 10 to 50. The descriptive statistics showed a mean school connectedness score of 25.29
with a standard deviation of .417. This is much lower than the mean school connectedness score
in the Theriot and Orme study of 33.2 with a standard deviation 6.5 (Theriot, Orme 2016). This
not only indicated lower levels of school connectedness in the participants of the present study,
but also a much less variable sample of school connectedness scores due to the lower standard
deviation. Additionally, the scale histogram was skewed to the right with higher proportions of
scores placed on the lower end of the range. This further indicates that the sample population has
lower levels of school connectedness.
Perceptions of Safety
The second survey scale, comprised of
6 questions, was meant to measure perceptions
of safety. Using the same 5 point Likert scale
as the last survey scale, students were asked if
they felt safe in various locations. This
included at school overall, in school hallways,
in the cafeteria, going to and from school, in school bathrooms, and their classrooms. The scale
score was summed the same as the last scale. The perceptions of safety scores had a possible
range of 6 to 30. The descriptive statistics showed a mean safety perceptions score of 22.32 and
standard deviation of .4366. Since the possible scores ranged from 6 to 30, the mean safety
perceptions score indicates higher feelings of safety within this sample population. The scale
01
0 0
21
0
4
21
6
9
3
89
14
17
32
6
23
7
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30Source: My December 2018 GHC Study
Safety Perceptions Scale Scores
histogram is also skewed to the left, with higher proportions of scores on the higher end of the
score range. This also supports the conclusion that this sample reported higher feelings of safety.
Experienced Violence
The third survey scale, comprised of 7
questions, measures the level of experienced
violence of the survey respondent. It asked the
participant questions of various types of violence
they may have personally experienced while at
school. This includes teasing, physical fights,
arguments, bullying, physical harm from a partner, racially based harassment, or physical threats.
The scores had a possible range from 7 to 40, in which higher score indicated higher levels of
experienced violence while at school. The sample population had a mean experienced violence
score of 10.11 with a standard deviation of 4.32. Compared to the sample population of the
Theriot and Orme study, which had a mean experienced violence scores of 12.8 and standard
deviation of 5.8, the present study’s sample indicated lower scores and less variability (Theriot,
Orme 2016). While the present study’s sample population had a lower mean experienced
violence score than the Theriot and Orme study, 12.8 is still within one standard deviation from
10.11, so it is not statistically significant. The scale histogram was dramatically skewed to the
right, with most scores on the lower end of the range. This is another indication that the current
study reported lower levels of experienced violence.
Witnessed Violence
27
19 19
10
6
3 42 1 1 1 1 0
20 1
3
0 0 0 0 0 0 0
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Source: My December 2018 GHC Study
Experienced Violence Scale Scores
The fourth survey section addressed
any witnessed violence. It included the
same types of violence as the last scale, but
it instead asked the participant whether
they had seen it happen to another student,
rather than to themselves. The scores
ranged from 7 to 40. The survey sample had a mean witnessed violence score of 20.95 with a
standard deviation of 8.35. While the Theriot and Orme study provided no descriptive statistics
for this scale, the mean witnessed violence score of the present study indicates that its population
had lower levels of witnessed violence. This is due to the mean score, 20.95, being lower than
the median score of the possible range (23.5). Overall, the scale histogram shows a moderate
skewness to the right. This suggests that similarly to the experienced violence scale, the current
study’s sample population reported lower levels of witnessed violence.
Opinions of SRO
This scale asked the participant 10
questions addressing their perceptions of their
SRO. This includes whether they liked having
the officer at school, felt safer with the SRO at
school, and whether they believed the officer
was helpful to students. Additionally, students
were asked if they believed the SRO treated all students fairly, does a good job, makes them feel
better about police, has a good relationship with students, and that other students like the officer.
They were also asked whether they felt their SRO does a good job of stopping violence and
5
4
5
7
9 9
7
0
8
4 4
2
3
7 7
1 1
3
0
2
0
2
1
2
0
1 1 1
0
1
0 0 0 0 0 0
1
0
1 1
11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49
Source: My December 2018 GHC Study
Witnessed Violence Scale Scores
2
6
15
32
1922
2 2
20 22 24 26 28 30 32 34
Source: My December 2018 GHC Study
SRO Perceptions Scale Score
preventing crime (sale of drugs or alcohol at school). There was a possible range of 10 to 50
points on this scale. The current study had a mean SRO perception score of 26.88 and a standard
deviation of 2.8. Compared to the Theriot and Orme study with a mean score of 33.8 and
standard deviation of 6.5, the present study’s sample population had less positive opinions of
their SRO (Theriot, Orme 2016). The scale histogram points to the same conclusion given its
higher concentration of scores at or under 30 (the median of the possible score range). Overall,
the sample population has reported less positive opinions of its SRO on campus.
Scale Correlations
Many relationships can be identified between the measured scales of school
connectedness, experienced violence, witnessed violence, and SRO perceptions. Previous
research, such as the study by Skiba and colleagues has indicated that school connection and
climate are more important than factors of delinquency or physical safety measures (such as
SROs) in predicting students’ overall feelings about school safety (Skiba et al. 2004).
Additionally, the study done by Theriot and Orme demonstrated that while students with more
positive attitudes toward SROs on campus felt safer, experiences with school violence, rather
than their interactions with SROs, had a more significant impact on their feelings of safety
(Theriot, Orme 2016). The study conducted by McDevitt and Panniello also found that
perceptions of safety were most impacted by the level of neighborhood crime, past victimization,
comfort reporting crimes, having a positive opinion of SRO, and gender (McDevitt, Panniello
2005). In order to identify the possible relationships established by previous researchers, various
correlations between the multiple scales were run using Excel technology.
Safety Perceptions
The correlations demonstrated that relationships similar to the ones found in previous
research were also present in this study’s sample population. For example, there was a slight
positive correlation of +.43 between the School Connectedness (SC) and Safety Perceptions (SP)
scores. Meanwhile, there was only a correlation of +.20 between the SRO Perceptions (SROP)
scores and the SP Scores. This is consistent with the results of the study done by Skiba and
colleagues, which also indicated that school climate and connection was more important than
security measures in predicting feelings of school safety (Skiba et al. 2004). This is also
supported by how there was only a slight negative correlation of -.28 between Experienced
Violence (EV) and SP scores and only a slight negative correlation of -.23 between Witnessed
Violence (EV) and SP scores. Just as in Skiba and colleagues’ study, school climate and
connection was more important in predicting safety perceptions than student delinquency.
Overall, SC scores had the strongest correlation with the SP scores.
Interactions and Opinions of SRO
SC Score SP Score EV Score WV Score SROP Score
SC Score 1.00
SP Score 0.43 1.00
EV Score -0.29 -0.28 1.00
WV Score -0.18 -0.23 0.65 1.00
SROP Score 0.19 0.20 -0.09 -0.09 1.00
slight positive slight negative
strong positive
Interactions Class/Presentation SC Score SP Score SROP Score
Interactions 1.00
Class/Presentation -0.06 1.00
SC Score -0.31 0.02 1.00
SP Score -0.01 -0.05 0.43 1.00
SROP Score -0.02 0.12 0.19 0.20 1.00
slight negative slight positive
In order to identify a relationship between interactions with the SROs and opinions of the
SROs, correlations were run between the two questions quantifying the student’s number of
interactions with the SRO and the SROP scale. Contrary to the results of the Theriot study of
SROs and school connectedness in 2016, the correlations indicated no significant relationship
between the number of interactions between students and SROs and more positive opinions of
the SROs (Theriot 2016). This is due to the correlation of -.02 between SROP and Interactions
and the correlation of +.12 between SROP and Classes/Presentations. However, the number of
interactions had a slight negative correlation of -.31 with the SC score, which is consistent with
the same Theriot study which indicated that increased interactions with SROs was related to
lower levels of school connectedness (Theriot 2016).
Multivariate Logistic Regression
Due to the only slight correlations between opinions of SROs and perceptions of safety
and school connectedness, further data analysis was done to identify any other significant
relationships. Two multivariate regressions were performed for the Safety Perceptions Scores
and the School Connectedness Scores.
Predicting Safety Perceptions
The most significant variable
in predicting Safety Perception (SP)
Scores was the SC Scores. The
multivariate regression resulted in the
SC scores having a coefficient .37 and p-value of 0% when predicting SP Scores. The coefficient
is the biggest out of the 3 variables, and it is most statistically significant due to its low p-value.
Safety Perceptions Score Regression
CoefficientsStandard Error t Stat P-value
Intercept 10.57 4.51 2.34 0.02
SC Score 0.37 0.10 3.78 0.00
EV Score -0.21 0.10 -2.13 0.04
SROP Score 0.16 0.14 1.16 0.25
The coefficient of the EV is much lower, with a slightly higher p-value of .04. While the EV
scores are still a statistically significant variable, its lower coefficient indicates it is less
important in predicting SP Scores. Once again, the data analysis is consistent with Skiba and
colleagues’ study, which concluded that school climate and connection was more important in
predicting safety perceptions than student delinquency, or EV scores in the present study (Skiba
et al. 2004). It is also consistent with the results of Goldstein and colleagues study where high
exposure to relational (verbal) aggression, which was included in the EV scale, made students
feel less safe at school (Goldstein et al. 2007). Additionally, the coefficient of the SROP was
notably lower than of SC, and its p-value of 25% indicated the variable was statistically
insignificant. This is consistent with the results of Theriot and Orme’s study that concluded
students’ experiences with school violence, rather than their interactions with SROs, had a more
significant impact on their feelings of safety (Theriot, Orme 2016).
Predicting School Connectedness
The most significant variable
in predicting SC scores was the SP
Scores. It had a coefficient of .35
and p-value of 0%. However, the
least impactful variable was the SROP Scores, since its coefficient of SROP was .15 and its p-
value was 28%, indicating it was statistically insignificant. This is contrary to the results of
Theriot’s study in 2016 concluded that SROs negatively affected school connectedness (Theriot
2016). The second most impactful variable was the EV Scores, with a coefficient of -.16 and p-
value of 10%. While the p-value was too big to label the variable statistically significant, it’s
negative impact on SC scores is consistent with the results of Goldstein and colleagues study
School Connectedness Score Regression
Coefficients Standard Error t Stat P-value
Intercept 15.08 4.22 3.58 0.00
SP Score 0.35 0.09 3.78 0.00
EV Score -0.16 0.09 -1.65 0.10
SROP Score 0.15 0.14 1.09 0.28
where high exposure to relational (verbal) aggression, which was included in the EV scale,
impacting students’ perceptions of their social climate negatively (Goldstein et al. 2007).
VI. Conclusion
The present study has provided further insight into the effects of SROs on safety
perceptions and feelings of school connectedness. The overall trends, correlations, and
multivariate regressions indicated that while SRO perceptions do impact feelings of safety and
school connectedness, experienced violence is instead the most impactful variable. The effects
on experienced violence is supported by the study conducted by Goldstein and Boyd’s study
where high exposure to relational (verbal) aggression, which was included in the EV scale,
impacting students’ perceptions of their social climate negatively and feel less safe at school.
Despite this, the number of interactions with the SRO still had an impact on feelings of school
connectedness due to its negative correlation. Overall, the results partially confirm the study’s
original hypothesis of SROs’ negative impact on school connectedness and positive impact on
feelings of safety because while there was only a slight positive relationship between SRO
opinions and feelings of safety, experienced violence was more impactful and significant. By
investigating these topics of school safety and connectedness in a student population that differs
ethnically and socioeconomically than previous research, a more diverse picture is captured of
student perceptions of SROs and their effects on safety perceptions and school connectedness.
Certain limitations to the present study may have contributed to the partial confirmation
of the study’s hypothesis. Only a small proportion of students (6%) indicated having any kind of
interaction with their SRO, so it may have led to the insignificant relationship between
interactions and feelings of safety. Additionally, this study surveyed a sample of 100 students,
so it is too small of a sample population to generalize to the much larger student body of 4,480
students. Despite its limitations, this study’s results contribute valuable insight into how student
opinions of SROs as well as their experiences at school affect their feelings of safety and school
connectedness. According the Benjamin Kutsyuruba and colleagues, a positive school climate,
safe school environment, and well-being of students are integral to meeting their academic,
emotional, and social needs (Kutsyuruba et al. 2015). By investigating the effects of safety
measures, such as School Resource Officers, researcher can further their understanding of how
their implementation can affect students experiences in school. The present study has provided a
new understanding of these effects of SROs on students of a new and diverse population that can
be applied to an extensive part of the student experience.
Bibliography
Bosworth, K., Ford, L., & Hernandaz, D. (2011). School Climate Factors Contributing to Student and
Faculty Perceptions of Safety in Select Arizona Schools*. Journal of School Health,81(4), 194-
201.
Brown, B. (2006). Controlling Crime and Delinquency in the Schools. Journal of School Violence,4(4),
105-125.
Brown, B., & Benedict, W. R. (2005). Classroom Cops, What Do the Students Think? A Case Study of
Student Perceptions of School Police and Security Officers Conducted in an Hispanic
Community. International Journal of Police Science & Management, 7(4), 264-285.
Corley, C. (2018). Do Police Officers In Schools Really Make Them Safer? Retrieved from
https://www.npr.org/2018/03/08/591753884/do-police-officers-in-schools-really-make-them-
safer
Cornell, D. G., & Loper, A. B. (1998). Assessment of violence and other high-risk behaviors with a
school survey. School Psychology Review,27(2), 317-331.
Goldstein, S. E., Young, A., & Boyd, C. (2007). Relational Aggression at School: Associations with
School Safety and Social Climate. Journal of Youth and Adolescence,37(6), 641-654.
Jackson, A. (2002). Police‐school resource officers’ and students’ perception of the police and
offending. Policing: An International Journal of Police Strategies & Management,25(3), 631-
650.
Kutsyuruba, B., Klinger, D. A., & Hussain, A. (2015). Context and Implications Document for:
Relationships among school climate, school safety, and student achievement and well-being: A
review of the literature. Review of Education,3(2), 136-137.
McDevitt, J., & Panniello, J. (2005). National Assessment of School Resource Officer Programs: Survey
of Students in Three Large New SRO Programs. U.S. Department of Justice.
Rosenblatt, J. A., & Furlong, M. J. (1997). Assessing the Reliability and Validity of Student Self-
Reports of Campus Violence. Journal of Youth and Adolescence,26(2), 187-202.
Skiba, R., Simmons, A. B., Peterson, R., Mckelvey, J., Forde, S., & Gallini, S. (2004). Beyond Guns,
Drugs and Gangs. Journal of School Violence,3(2-3), 149-171.
Theriot, M. T. (2016). The Impact of School Resource Officer Interaction on Students’ Feelings About
School and School Police. Crime & Delinquency,62(4), 446-469.
Theriot, M. T., & Orme, J. G. (2014). School Resource Officers and Students’ Feelings of Safety at
School. Youth Violence and Juvenile Justice,14(2), 130-146.
Varjas, K., Henrich, C. C., & Meyers, J. (2009). Urban Middle School Students Perceptions of Bullying,
Cyberbullying, and School Safety. Journal of School Violence, 8(2), 159-176.
Addendum
My December 2018 GHC Study